Priority Paper The metabolic pathway of visual ... - Wiley Online Library

Eur. J. Biochem. 257, 5222527 (1998)  FEBS 1998

Priority Paper The metabolic pathway of visual pigment chromophore formation in Drosophila melanogaster All-trans (3S)-3-hydroxyretinal is formed from all-trans retinal via (3R)-3-hydroxyretinal in the dark Takaharu SEKI 1 , Kunio ISONO 2, Kaoru OZAKI 2, Yasuo TSUKAHARA 2, Yuko SHIBATA-KATSUTA 3, Masayoshi ITO 3, Toshiaki IRIE 4 and Masanao KATAGIRI 5 1 2 3 4 5

Division of Health Science, Osaka Kyoiku University, Osaka, Japan Graduate School of Information Sciences, Tohoku University, Sendai, Japan Kobe Pharmaceutical University, Kobe, Japan Osaka Meijo Women’s College, Osaka, Japan Division of Natural Science, Osaka Kyoiku University, Osaka, Japan

(Received 20 June/11 August 1998) 2 EJB 98 0796/1

Carotenoid-depleted fruit flies, Drosophila melanogaster, were reared on yeast/glucose medium containing lipid-depleted white corn grits and cholesterol. After rearing for more than a year, the yield of flies remained constant and the content of 3-hydroxyretinal in a head was three logarithmic units less than that of normal flies reared on medium containing yellow corn grits. When all-trans retinal was supplied as the sole source of retinoids, the flies formed and accumulated all-trans 3-hydroxyretinal in the dark. To examine the metabolic pathway to produce (3S)-3-hydroxyretinal in Drosophila, all-trans retinal was supplemented for two hours to carotenoid-depleted flies in the dark, and the subsequent changes in the composition of 3-hydroxyretinal enantiomers were analyzed using a chiral column on HPLC. The results indicated initial formation of (3R)-3-hydroxyretinal followed by isomerization into the 3S enantiomer. In another set of experiments, the membrane fraction was obtained from the head homogenate of retinoid-depleted flies and an in vitro assay of 3-hydroxyretinal formation from retinal was performed. The 3-hydroxyretinal produced was the 3R enantiomer, supporting the result obtained from the in vivo experiment whereby (3S)-3-hydroxyretinal is produced from retinal via (3R)-3-hydroxyretinal. Addition of NADPH enhanced 3-hydroxyretinal formation and the presence of carbon monoxide inhibited it, suggesting that hydroxylation at the C3 position of retinal occurred via the monooxygenase activity of cytochrome P-450. Keywords : chiral discrimination ; chromophore; cytochrome P-450 ; Drosophila melanogaster; 3-hydroxyretinal.

Absorption of a photon by a visual pigment is the initial step of vision, and the visual pigment is composed of the 11-cis isomer of a retinal congener (chromophore) bound to a protein moiety (opsin). In the animal kingdom, five retinal congeners, retinal, 3,4-didehydroretinal, (3R)-3-hydroxyretinal, (3S)-3-hydroxyretinal and (4R)-4-hydroxyretinal, have been known to be used as visual pigment chromophores, and three of them are used in the class Insecta; retinal, (3R)-3-hydroxyretinal and (3S)3-hydroxyretinal [1, 2]. Among these, retinal is considered to be the oldest [1, 3, 4], since it is the only chromophore used in the insect orders that have been extant since the Carboniferous peCorrespondence to T. Seki, Division of Health Science, Osaka Kyoiku University, Asahigaoka, Kashiwara, Osaka 582-8582, Japan Phone: 181 729 78 3611. Fax: 181 729 78 3554. E-mail: [email protected] Abbreviations. DWC, lipid-depleted white corn; YC, yellow corn ; lutein A, (3R, 3′R, 6′R)-β,ε-carotene-3,3′-diol; β-cryptoxanthin, β-carotene-3-ol; zeaxanthin, β,β’-carotene-3,3′-diol.

riod [2]. In contrast to retinal, 3-hydroxyretinal has a chiral center at the C3 position in the β-ring, therefore there are two possible stereo isomers, (3R)-3-hydroxyretinal and (3S)-3-hydroxyretinal. Among insects that use 3-hydroxyretinal, all species examined so far in the orders Odonata, Hemiptera, Coleoptera, Neuroptera and Lepidoptera, and in suborders Nematocera and Brachycera in the order Diptera, have been found to use only the 3R enantiomer of 3-hydroxyretinal [1, 2]. As the 3-hydroxyxanthophylls produced by plants, algae and bacteria have only the (3R)-3-hydroxy-β-ring [5, 6], the (3R)-3-hydroxyretinal used by insects has been supposed to be produced directly from 3hydroxyxanthophylls [123], in a similar way to which retinal is produced from carotene in mammals [7, 8]. It is considered that (3R)-3-hydroxyretinal has arisen as the chromophore of some insects since around the end of Carboniferous period [2]. The (3S)-3-hydroxyretinal chromophore has been found to be used by all species examined so far in the suborder Cyclorrhapha (higher flies) in Diptera. In these flies 11-cis 3-hydroxyretinal is exclusively (3S)-enantiomer while all-trans 3-hydroxyretinal is

523

Seki et al. (Eur. J. Biochem. 257)

Table 1. Carotenoid compositions in the yellow corn, white corn and lipid-depleted white corn analyzed by HPLC. Lutein, zeaxanthin and β-cryptoxanthin (β-Crypt.) were analyzed by a normal phase column, and the solvent-front fraction containing A- and β-carotene was re-analyzed with a reverse phase column. Lutein

Zeaxanthin

β-Crypt.

A-Carotene

β-Carotene

Sum

1078.00 3.28 0.06

387.00 2.80 2

20.00 2 2

22.00 2 2

3401.00 12.90 0.22

pmol/g Y-corn W-corn Lipid-depleted W-corn

1894.00 6.84 0.15

a mixture of both the 3R and 3S enantiomers [1, 2]. These results suggest that the higher flies, which is a new insect group appearing since the Jurassic, have acquired a novel metabolic pathway to produce (3S)-3-hydroxyretinal. In Drosophila melanogaster (Cyclorrhapha, Diptera), retinoids and carotenoids serve as precursors of the chromophore of visual pigment [9], but the light conditions required for these pathways from each precursor are different. Starting from carotenoids, the fly can form 11-cis 3-hydroxyretinal in the dark [10212], but if retinoids are used, the presence of light is obligatory [12] as shown also for the blow fly, Calliphora erythrocephara [13]. Therefore, when all-trans retinal is used as the starting material in the absence of light, all-trans 3-hydroxyretinal accumulates in the fly head [12] without the following isomerization to 11-cis 3-hydroxyretinal. In the present paper, the chirality of 3-hydroxyretinal produced in the dark from all-trans retinal in vivo and in vitro is investigated as the first step to elucidating the formation of (3S)-3-hydroxyretinal in Drosophila. MATERIALS AND METHODS Fly. A mutant white (w) stock of D. melanogaster (a spontaneous mutation of w in Canton-S) was used throughout the study. They were grown either under 12 h light/12 h dark light cycles or continuous dark in an incubator at 25°C on 25 ml medium containing dry yeast (Oriental Yeast Co.), cornmeal (Honen Oil Production Co.) and glucose [12] in a vial (4 cm diameter312 cm height). Two different types of cornmeal were used in this study for controlling the carotenoid content in the media : yellow corn (YC) grits (60 g in 1 l medium) used as carotenoid-rich medium for rearing control flies, and lipid-depleted white corn (DWC; 60 g in 1 l medium) for rearing carotenoid-depleted flies. The latter medium was supplemented with cholesterol (300 mg in 1 l medium) which was lost by the lipid extraction procedure. Table 1 shows the composition and content of carotenoids in grits of yellow, normal white and lipid-depleted white corn. Adult flies eclosed in the vials were transferred to new vials containing the same medium and maintained until the experiments as described later. About 1000 flies/vial were obtained, even with the DWC medium. Lipid depletion of white corn grits. To a suspension of 200 g white corn grits in 100 ml water, were added 100 ml isopropanol and 200 ml dichloromethane, and the suspension was homogenized for 5 min with a high speed homogenizer (Physcotron NS-50; NITI-ON) equipped with a generator shaft with a diameter of 20 mm. The mixture was filtered under vacuum on a Buchner funnel with a paper filter (Advantec Toyo, filter paper no. 2). The funnel was covered with an air-tight polychlorvinylidene sheet (Kureha Chemical Kogyo Co.). The lipid-extraction procedure was repeated three times, and the final white corn powder was suspended in methanol and filtered under vacuum on the Buchner funnel to complete dryness.

Retinoids and carotenoids. All-trans retinal was purified from concentrated retinal (a gift from Dr Y. Kito). Following TLC with a silica gel 60 TLC plate (200 mm3200 mm; Merck), the all-trans retinal fraction was collected by preparative HPLC equipped with a 10-mm diameter3250 mm column of silica gel (particle size, 5 µm; LichroSorb SI-60, Merck). After two steps of purification by preparative HPLC, the purity of all-trans retinal exceeded 99%. Absorption spectra were measured with a spectrophotometer (U-3200, Hitachi Co., Ltd.). The concentration of purified all-trans retinal in ethanol was calculated using the molar absorption coefficient value (εmax) 42900 M21 cm21 [14] at the absorption maximum (λmax). Racemic all-trans 3-hydroxyretinal was synthesized as reported [15]. After exposure to white light at wavelengths above 420 nm in ethanol, 11-cis and all-trans isomers were collected using HPLC, and the concentration of each isomer was calculated from its λmax in ethanol, using εmax values of 44000 M21 cm21 and 21 000 M21 cm 21 [15], respectively. Lutein A, (3R, 3′R)-zeaxanthin and (3R)-β-cryptoxanthin were gifts from Drs T. Maoka and M. Tsushima, and the concentrations in ethanol of purified samples were calculated using ε max values 144 800 M21 cm 21 and 144 300 M21 cm 21 [16] for lutein and zeaxanthin, respectively. The concentrations of βcarotene (Sigma Chem. Corp.) and A-carotene (Wako Pure Chemical Industries, Ltd.) in n-hexane were calculated using εmax values 139 500 M21 cm21 and 146 000 M21 cm 21 [17], respectively. HPLC analysis. The HPLC system consisted of a pump (Hitachi 655), two absorbance detectors (875-UV & UV-970; Japan Spectroscopic Co., Ltd.) and an integrator (SIC-Labchart 80; System Instruments Co., Ltd.) essentially as described previously [18]. For analyses of retinoids and xanthophylls, a 6-mm diameter3150 mm column of silica gel (particle size, 3 µm; YMC-Pack A012-3 S-3 SIL, Yamamura Chemical Industry Co., Ltd.) was used at a flow rate of 2 ml/min. The eluents used were 5% tert-butylmethylether, 0.04% ethanol and 25% benzene in n-hexane, for separation of retinal isomers, and 25% tert-butylmethylether, 0.3% ethanol and 25% benzene in n-hexane for 3-hydroxyretinal isomers and xanthophylls. Optical resolution of all-trans 3-hydroxyretinal using a chiral column (ChiraSpher, 4-mm diameter3250 mm, Merck) and calculation of the ratio of 3R enantiomer to the sum of both enantiomers (3R ratio) was performed as described previously [1]. For analysis of carotene, the HPLC system consisted of a Waters 600 control system (Waters Div. of Millipore), a reverse-phase column (Cosmosil 5C18, 4.6-mm diameter3250 mm ; Nacalai Tesque Co., Ltd.), an absorbance detector (UV-875, Japan Spectroscopic) and an integrator (D-2000, Hitachi). The eluent composition was 20% dichloromethane and 10% methanol in acetonitrile [19], at a flow rate of 1.5 ml/min. For quantification of retinoids and carotenoids by HPLC, each purified standard sample, whose concentration had been measured by spectrophotometry, was applied quantitatively to

524


Table 2. Contents of 3-hydroxyretinal isomers in the heads of flies reared on the media containing YC or DWC. On DWC medium, flies reared for different generations after transfer from the YC medium have been analyzed n, number of fly heads analyzed by HPLC. n

3-Hydroxyretinal all-trans

11-cis

9-cis

Sum

fmol/head YC-medium YC-medium DWC-medium DWC-medium DWC-medium

One generation 6 months 12 months

870 200

146.00 118.00

175.00 304.00

8.00 9.00

329.00 431.00

4 240 9 290 18 000

1.40 0.11 0.18

1.30 0.11 0.29

0.00 0.00 0.00

2.70 0.22 0.47

the HPLC and the peak area of 1 pmol authentic sample was calculated from the chromatogram. All-trans retinal supplementation of the carotenoid-depleted Drosophila in the dark. Adult flies reared on DWC medium were starved in an empty vial containing wet paper for one day and transferred to a supplementation vial in which a mixture of 0.1 mg all-trans retinal and 100 mg DWC powder was spread on DWC medium. For pulse supplementation of retinal, flies were fed with all-trans retinal for 2 h, then transferred to new DWC media and collected by quick-freezing in liquid nitrogen at 0, 2, 6, 12 and 24 h after transfer. The fly heads were isolated by vigorous vibration and sieving as described elsewhere [12]. All the procedures were carried out in the dark or under the dim red light if necessary for manipulations. The number of fly heads was estimated from a unit head mass of 0.1 mg. Retinoids in the heads were extracted by the formaldehyde method [1]. Assay for 3-hydroxyretinal formation in vitro. Heads of D. melanogaster reared on DWC medium for successive generations were homogenized in 0.1 M Tris/HCl, pH 7.4 (Tris buffer), with a Physcotron equipped with a micro shaft of 7 mm diameter, for 1 min on ice. The homogenate was centrifuged at 12 000 g for 20 min and the precipitate (membrane fraction) was suspended in Tris buffer. For investigation of the effects of CO and NADPH on 3hydroxyretinal formation, the suspension of membrane fraction obtained from 3000 heads was divided into two portions. One portion (1CO) was bubbled with CO gas (GL Sciences Inc.) for 2 min and divided into two 1-ml aliquots in glass tubes with screw caps. The control portion (2CO) was also divided into two 1-ml aliquots, without CO bubbling. To each suspension was added 1.1 mM all-trans retinal dissolved in 10 µl ethanol. 10 mM NADPH (Sigma) in 100 µl Tris buffer was then added to one aliquot of each treatment (1CO and 2CO). The mixtures were incubated at room temperature (21°C) for 25 min, during which CO gas was twice flushed on the 1CO samples. The reaction was stopped by addition of 2 ml formaldehyde and the retinoids in the mixture were extracted [1]. To analyze the effect of the substrate concentration, the membrane fraction from about 5000 heads of D. melanogaster reared on DWC medium for months were suspended in Tris buffer and divided into two aliquots, with or without 1 mM NADPH. All-trans retinal was added to subaliquots of each at concentrations of 0.9 µM or 9 µM and, after the appropriate time intervals, retinoids in portions of a mixture were extracted by the formaldehyde method and analyzed by HPLC [1]. All experiments were performed under red light of wavelengths exceeding 610 nm.

Fig. 1. Elution profiles of 3-hydroxyretinal isomers extracted from (a) 870 heads of control flies reared on medium containing yellow corn and (b) extract from 9290 heads of carotenoid-depleted flies reared for 6 months on medium containing lipid-depleted white corn. Substances eluted between 7.5 min and 11 min were collected from both (a) and (b), and a small aliquot of that from (a) and all from (b) were eluted together on the same column (c). Scale bar indicates absorbance at 360 nm: 3.231023 for (a); 231024 for (b) and (c).

RESULTS Retinoid content in the heads of D. melanogaster reared on DWC medium. After transfer from YC medium to DWC medium, the amount of 3-hydroxyretinal in the fly head decreased by more than two logarithmic units after one generation (Table 2) ; the flies raised solely on DWC medium was bred on it continuously. Fig. 1b is an HPLC chromatogram of lipid-soluble substances extracted from 9290 heads of D. melanogaster reared on DWC medium for about 6 months. The two small peaks at about 8 min and 10 min have the same retention time as alltrans and 11-cis 3-hydroxyretinal, respectively, extracted from control flies raised on YC medium (Fig. 1a), as confirmed by co-chromatography (Fig. 1c). However the 8-min peak had a higher absorbance at 330 nm than that at 360 nm, indicating the presence of some substance other than 3-hydroxyretinal. The peak area of the 8-min fraction, however, has been used to quantify all-trans 3-hydroxyretinal in the heads of flies reared on


Fig. 2. Change of the chirality of all-trans 3-hydroxyretinal in the heads of carotenoid-depleted flies supplemented with all-trans retinal for 2 h in the dark, followed by incubation in the dark on retinalfree DWC medium. (A) Chromatograms of all-trans 3-hydroxyretinal (chiral column HPLC) at times 0 (D-0), 2 (D-2), 6 (D-6), 12 (D-12) and 24 (D-24) h in the dark after the 2 h supplementaion of all-trans retinal (arbitrary absorption units at 360 nm). (B) Time-dependent changes in the amount of total all-trans 3-hydroxyretinal (1), all-trans (3R)-3-hydroxyretinal (m) and all-trans (3S)-3-hydroxyretinal (.) in fmol/head (left ordinate); and of the ratio of all-trans (3R)-3-hydroxyretinal to total 3-hydroxyretinal (3R ratio ; broken line, right ordinate) after supplementation with all-trans retinal for 2 h (bar on the abscissa).

DWC medium (Table 2). In another analysis using 18 000 heads of flies reared on DWC medium for about one year, the HPLC fraction including the two peaks at 8 min and 10 min was collected and reduced with sodium borohydride. Re-chromatography by HPLC shows peaks corresponding to all-trans and 11cis 3-hydroxyretinol isomers (data not shown), indicating that all-trans and 11-cis 3-hydroxyretinal had been present in the collected fraction. The results (Table 2) indicate that the total amounts of 3-hydroxyretinal in the heads of these flies reared

525

Fig. 3. Effects of NADPH and CO on the formation of 3-hydroxyretinal in vitro from all-trans retinal using the membrane fraction of fly heads. (A) A suspension of membrane fraction obtained from about 3000 heads of D. melanogaster reared on DWC media was divided into four aliquots, and incubated with 11 µM all-trans retinal for 25 min by addition of 1 mM NADPH (a and c) or without NADPH (b and d), and by bubbling with CO gas (c and d) or without CO bubbling (a and b). As the standard chromatogram, the all-trans and 11-cis 3-hydroxyretinal isomers extracted from heads of control flies reared on YC medium are shown (e). The bar indicates the absorbance of 360 nm at 231024 for (a2d), and at 3.231023 for (e). (B) Identification of the all-trans (3R)-3hydroxyretinal in the 8-min fraction [(a) in Fig. 3A] by the chiral column HPLC. The main peak fraction at 28.5 min in the upper chromatogram was collected and re-chromatographed with the racemic all-trans 3-hydroxyretinal (lower) to confirm the main peak to be the 3R enantiomer. The bar indicates the absorbance of 360 nm at 231024.

on DWC medium, are three logarithmic units less than in control flies reared on YC medium. Retinoid composition in the head of D. melanogaster reared on DWC medium supplemented with all-trans retinal. Flies reared on DWC medium supplemented with all-trans retinal in the dark produce only all-trans 3-hydroxyretinal [12]. When the flies were collected just after supplementation with all-trans retinal for 24 h in the dark, the chirality of all-trans 3-hydroxyretinal produced was a mixture of 3R and 3S enantiomers with the

526


the inhibitory effect by CO (see Discussion). The peak substance at a retention time of 8 min has been produced in the presence of NADPH without CO, but suppressed in the presence of CO (Fig. 3A). The peak fraction at 8 min was collected and re-chromatographed with the chiral column. The 8-min fraction contains all-trans 3-hydroxyretinal and it is exclusively the 3R enantiomer (Fig. 3 B). These results support the hypothesis that 3hydroxyretinal is produced from retinal by the enzymatic activity of cytochrome P-450 in the membrane fraction, and indicate that the absolute structure of 3-hydroxyretinal produced is 3R. Addition of NADPH augmented the initial formation rate of 3-hydroxyretinal in vitro (Fig. 4). Without added NADPH, the amount of 3-hydroxyretinal remained constant after initial formation, probably due to the consumption of endogenous NADPH, but in the presence of added NADPH, the amount of 3-hydroxyretinal decreased after reaching a maximum. The initial rate of 3-hydroxyretinal formation depends on the concentration of substrate, all-trans retinal (Fig. 4), suggesting that the 3hydroxyretinal formation is catalyzed by the enzymatic activity. Fig. 4. Time courses of 3-hydroxyretinal formation in vitro from alltrans retinal using the membrane fraction of fly heads. About 5000 heads were centrifuged following homogenization. To one fraction was added 1 mM NADPH (filled symbols), and to the other, added only the buffer (open symbols). All-trans retinal was added to each aliquot at 9 µM (circles) or 0.9 µM (triangles), and portions of the mixture were drawn out to analyze the 3-hydroxyretinal by HPLC at the time intervals shown on the abscissa. The 8-min fraction on the HPLC chromatogram (Fig. 3A) was regarded as the all-trans 3-hydroxyretinal, and the peak area of 8-min fraction was divided by the peak area of the 4.9-min fraction (Figs 1 and 3A) to normalize the amount of 3-hydroxyretinal with the membrane content in each aliquot.

3R ratio at 0.36. However, flies collected after 24 h incubation in the dark on retinal-free DWC medium, following 24 h supplementation with all-trans retinal, the 3R ratio was 0.20; that is, the ratio of 3S enantiomer increased during dark incubation. In a different experiment, flies supplemented with all-trans retinal for 2 h in the dark were transferred to new DWC medium, and the chirality of 3-hydroxyretinal subsequently produced at 0, 2, 6, 12 and 24 h in the dark was analyzed. Surprisingly, the initial product was predominantly the 3R enantiomer with the 3R ratio around 0.74. During incubation in the dark, the 3R ratio decreased to about 0.5 after 24 h (Fig. 2). From the total amount of all-trans 3-hydroxyretinal and the 3R ratio, the respective amounts of 3R and 3S enantiomers were calculated and plotted (Fig. 2 B). Extrapolation of the curves back to zero on the ordinate suggests that 3-hydroxyretinal formation began about 1 h after supplementation with all-trans retinal. In the initial phase of 3-hydroxyretinal formation, the formation rate of the 3R enantiomer exceeded that of the 3S enantiomer, but after 12 h, when the formation rate of total 3-hydroxyretinal decreased (presumably due to the consumption of retinal), the amount of 3R enantiomer decreased while that of the 3S enantiomer increased. These results suggested strongly that, initially, (3R)-3-hydroxyretinal was produced from all-trans retinal, then this was isomerized to the 3S enantiomer. Formation of 3-hydroxyretinal in vitro from retinal using the membrane fraction obtained from the head homogenate. As 3-hydroxyretinal is the product of a monooxygenase reaction upon retinal, this suggests the involvement of a cytochrome P450. To confirm this hypothesis, the membrane fraction obtained from the head homogenate of flies reared on DWC medium was assayed for the production of 3-hydroxyretinal from all-trans retinal upon the stimulatory effect by addition of NADPH and

DISCUSSION As neither retinoids nor carotenoids are essential for the survival of D. melanogaster [12, 24], it has become established practice to rear the fly on the medium deprived of retinoids and carotenoids. To obtain abundant carotenoid-depleted flies, we initially used a yeast/glucose medium supplemented with white corn grits as the vitamin-A-deprived medium, but white corn grits were subsequently found to still include about 0.4% of the carotenoids contained in YC grits (Table 1) supporting the results presented by Stark [25]. Lipids in the white corn grits were, therefore, extracted with dichloromethane/alcohol/water, and carotenoids were thus depleted to about 0.2 pmol/g (Table 1). Using DWC medium supplemented with cholesterol, about 1000 flies could be obtained from one vial containing about 1.5 g DWC (0.3 pmol carotenoids). On the assumption that all carotenoids in the vial was used by 1000 flies, it is calculated that the carotenoid content is 0.3 fmol/fly head. This value agrees well with the retinoid content of flies bred over successive generations on the DWC medium (0.220.5 fmol/head; Table 2). Although the estimation of the retinoid content in the head of these flies is possibly overestimated, the presence of 3-hydroxyretinal at nearly 0.1% of the amount of control flies is evident. However, this does not cause a problem for the HPLC analysis of exogenous retinoids in the heads of flies. In-the-dark supplementation of flies reared on DWC medium with all-trans retinal allowed us to isolate the process of hydroxylation at the C3 position of retinal in vivo. The results of pulse supplementation with all-trans retinal for 2 h (Fig. 2) indicated initial formation of (3R)-3-hydroxyretinal, followed by isomerization to the 3S enantiomer. From the characteristics of monooxygenase activity on retinal, we speculated that the enzymatic activity of the cytochrome P-450 system, which oxidates a variety of hydrophobic substances including retinoids in mammals [20222], might be involved in the hydroxylation of retinal to 3-hydroxyretinal in fly heads. A microsomal electron transport system containing a NADPH-cytochrome P-450 reductase requires NADPH as the source of reducing equivalent, and the terminal enzyme, cytochrome P-450, shows the monooxygenase activity which is inhibited by CO because CO binds the six coordinate of ferric heme in cytochrome P-450 [23]. Thus, the membrane fraction of the fly-head homogenate was assayed for 3hydroxyretinal formation with or without addition of NADPH, and in the presence or absence of CO. It was shown that the reaction was enhanced by addition of NADPH and inhibited in


the presence of CO (Fig. 3 A), supporting the cytochrome P-450 hypothesis. The absolute structure of 3-hydroxyretinal produced was exclusively the 3R enantiomer (Fig. 3 B), suggesting that the cyclorrhaphae chromophore, (3S)-3-hydroxyretinal, is produced from retinal via (3R)-3-hydroxyretinal. The conclusions from the present work suggest three further lines of research to clarify the metabolic pathway of visual pigment chromophore formation in Drosophila. Firstly, the isomerization process from the 3R to the 3S enantiomer requires confirmation of the presence of a racemase. In an in vitro experiment using head homogenate before centrifugation, both (3R)3-hydroxyretinal and (3S)-3-hydroxyretinal were produced (our unpublished data), suggesting the presence of isomerase activity in the water-soluble fraction. Secondly, it is necessary to elucidate the pathway resulting in 11-cis (3S)-3-hydroxyretinal formation. Under the experimental conditions of this paper, light inevitably produces the 11-cis-isomer. The present results suggest that a substrate for the putative photo-isomerase is all-trans (3S)-3-hydroxyretinal. Thirdly, it is necessary to clarify the metabolic pathway starting from carotenoids [10, 12, 26] which are considered to be the native precursor for visual pigment chromophore of Drosophila. 11-cis 3-Hydroxyretinal is produced without light [10, 12], but the pathway is yet a problem to be clarified. Another interesting question raised by the present results is whether or not insects using only (3R)-3-hydroxyretinal possess the same cytochrome P-450 hydroxylation activity as Drosophila. Further study of these problems will help to clarify the biological significance of the diversity of visual pigment chromophores across the animal kingdom. We are grateful to Drs T. Maoka and M. Tsushima (Kyoto Pharmaceutical University, Kyoto, Japan) for their supply of authentic 3-hydroxy xanthophylls. Thanks are also due to Dr K. Vogt (Universität Freiburg, Germany) for discussion and to Dr I. Gleadall (Ohu University, Fukushima, Japan) for comments on the manuscript.

REFERENCES 1. Seki, T., Isono, K., Ito, M. & Katsuta, Y. (1994) Flies in the group Cyclorrhapha use (3S)-3-hydroxyretinal as a unique visual pigment chromophore, Eur. J. Biochem. 226, 6912696. 2. Seki, T. & Vogt, K. (1998) Evolutionary aspects of the diversity of visual pigment chromophores in the class Insecta, Comp. Biochem. Physiol. 119B, 53264. 3. Vogt, K. (1987) Chromophore of insect visual pigments, Photobiochem. Photobiophys. Suppl., 2732296. 4. Smith, W. C. & Goldsmith, T. H. (1990) Phyletic aspects of the distribution of 3-hydroxyretinal in the class Insecta, J. Mol. Evol. 30, 72284. 5. Maoka, T., Arai, A., Shimizu, M. & Matsuno, T. (1986) The first isolation of enantiomeric and meso-zeaxanthin in nature, Comp. Biochem. Physiol. 83B, 1212124. 6. Maoka, T. (1990) Studies on the separation of carotenoid stereoisomers, Ph. D. Thesis, Kyoto Pharmaceutical University, Kyoto, Japan. 7. Goodman, D. S. & Huang, H. S. (1965) Biosynthesis of vitamin A with rat intestinal enzymes, Science 149, 8792880.

527

8. Olson, J. A. & Hayaishi, O. (1965) The enzymatic cleavage of βcarotene into vitamin A by soluble enzyme of rat liver and intestine, Proc. Natl Acad. Sci. USA 54, 136421370. 9. Lee, R. D., Thomas, C. F., Marietta, R. G. & Stark, W. S. (1996) Vitamin A, visual pigments, and visual receptors in Drosophila, Microsc. Res. Tech. 35, 4182430. 10. Seki, T., Fujishita, S., Ito, M., Matsuoka, N., Kobayashi, C. & Tsukida, K. (1986) A fly, Drosophila melanogaster, forms 11-cis 3-hydroxyretinal in the dark, Vision Res. 26, 2552262. 11. Isono, K., Tsukahara, Y. & Tanimura, T. (1984) Rhodopsin synthesis in the compound eye by supplement of retinoids and carotenoids to vitamin A-deficient fly, Zool. Sci. 1, 861. 12. Isono, K., Tanimura, T., Oda, Y. & Tsukahara, Y. (1988) Dependency on light and vitamin A derivatives of the biogenesis of 3hydroxyretinal and visual pigment in the compound eyes of Drosophila melanogaster, J. Gen. Physiol. 92, 5872600. 13. Schwemer, J. (1984) Renewal of visual pigment in photoreceptors of the blowfly, J. Comp. Physiol. 154, 5352547. 14. Liu, R. S. & Asato, A. E. (1984) Photochemistry and synthesis of stereoisomers of vitamin A, Tetrahedron 40, 193121969. 15. Ito, M., Matsuoka, N., Tsukida, K. & Seki, T. (1988) Retinoids and related compounds. X. Synthesis of geometrical isomers of (6 )2- and (6 )-3-hydroxyretinals and identification of the chromophore of the fly visual pigment, Chem. & Pharm. Bull. (Tokyo) 36, 78286. 16. Davis, B. H. (1976) Carotenoids, in Chemistry and biochemistry of plant pigments (Goodman, T. W., ed.) vol. 2, chapt. 4, pp. 1492 165, Academic Press, London, New York, San Francisco. 17. Tsukida, K. (1979) Vitamin A, in Biochemistry databook (Biochemical Society of Japan, ed.) vol. 1, pp. 127421288, Tokyo-KagakuDojin, Tokyo, Japan. 18. Seki, T., Fujishita, S., Ito, M., Matsuoka, N. & Tsukida, K. (1987) Retinoid composition in the compound eyes of insects, Exp. Biol. 47, 952103. 19. Olmedilla, B., Granado, F., Rojas-Hidalgo, E. & Blanco, I. (1990) A rapid separation of ten carotenoids, three retinoids, alpha-tocopherol and D-alpha-tocopherol acetate by high performance liquid chromatography and its application to serum and vegetable samples, J. Liq. Chromatogr. 13, 145521483. 20. Roberts E. S., Vaz, A. D. & Coon, M. J. (1992) Role of isozymes of rabbit microsomal cytochrome P-450 in the metabolism of retinoic acid, retinol, and retinal, Mol. Pharmacol. 41, 4272 433. 21. Martini, R. & Murray, M. (1993) Participation of P450 3A enzymes in rat hepatic microsomal retinoic acid 4-hydroxylation, Arch. Biochem. Biophys. 303, 57266. 22. Raner, G. M., Vaz, A. D. & Coon, M. J. (1996) Metabolism of alltrans, 9-cis, and 13-cis isomers of retinal by purified isozymes of microsomal cytochrome P450 and mechanism-based inhibition of retinoid oxidation by citral, Mol. Pharmacol. 49, 5152522. 23. Omura, T. (1993) Introduction: History of cytochrome P-450, in Cytochrome P-450 (Omura, T., Ishimura, Y. & Fujii-Kuriyama., T., eds) 2nd edn, pp. 1215, Kodansha Ltd., Tokyo, Japan. 24. Stark, W. S. (1997) Comparative biology of retinoid deprivation and replacement in flies and rodents, in Degenerative retinal diseases (LaVail, M. M., Hollyfield, J. G. & Anderson, R. E., eds) pp. 1352143, Plenum Press, New York. 25. Stark, W. S. (1977) Diet, vitamin A and vision in Drosophila, Drosophila Information Service 52, 47. 26. Vogt, K. (1983) Is the fly visual pigment a rhodopsin? Z. Naturforsch. 38c, 3292333.